The major purpose of the proposed research is to explore aspects of the development of prosocial behavior in children, and the manner in which prosocial behavior is determined. Research on development will focus on "natural socialization," on learning that results from "experience" in contrast to instruction by socializers. The effects of two types of experience, participation in prosocial behavior--of a kind that places direct responsibility on children for others welfare, and involves interaction with others, the supervision of younger children by older ones--and children teaching other children, will be experimentally explored. Under certain conditions, when feelings of responsibility and of benefiting others is associated with a sense of effectiveness or importance or satisfaction, both activities are expected to increase prosocial behavior. The effects of variation in some of these conditions on both cognition (e.g. self perception) and prosocial behavior will be explored, so that the cognitive bases of changes in prosocial behavior can be evaluated. Other research will attempt to change negative patterns of children's interactions with their peers to more positive ones. An extensive role playing procedure will be employed. Existing research findings suggest that unpopular children have adjustment problems in later life. The major purpose of the above research is to interfere with such a direction in children's personality development, by enabling them to improve their relationship with peers. Finally, research will be conducted to explore the interaction between personality characteristics and the nature of situations in determining whether children will or will not behave prosocially. A specific model is proposed which focuses on the significance of a person's motives (personal goals), the activating potential of the situation for various goals, and on goal conflict and its resolution as determinants of behavior. Competencies and other characteristics are also considered. Measures of personal goals and of situational activation will be developed, and then used in research to predict behavior.